No abstract

Bearing state recognition, especially under variable working conditions, has the problems of low reusability of monitoring data, low state recognition accuracy and low generalization ability of the model. The feature-based transfer learning method can solve the above problems, but it needs to rely on signal processing knowledge and expert diagnosis experience to obtain the cross-characteristics of different working conditions data in advance. Therefore, this paper proposes an improved balanced distribution adaptation (BDA), named multi-core balanced distribution adaptation (MBDA). This method constructs a weighted mixed kernel function to map different working conditions data to a unified feature space. It does not need to obtain the cross-characteristics of different working conditions data in advance, which simplifies the data processing and meet end-to-end state recognition in practical applications. At the same time, MBDA adopts the A–Distance algorithm to estimate the balance factor of the distribution and the balance factor of the kernel function, which not only effectively reduces the distribution difference between different working conditions data, but also improves efficiency. Further, feature self-learning and rolling bearing state recognition are realized by the stacked autoencoder (SAE) neural network with classification function. The experimental results show that compared with other algorithms, the proposed method effectively improves the transfer learning performance and can accurately identify the bearing state under different working conditions.

In some developed countries, the automatic vehicle recognition is a quite mature technology. This paper applies the multi-classification method based on Support Vector Machine (SVM) to vehicle recognition. Support vector machine, appeared recently, is a new theory and technology in the filed of pattern recognition and has shown excellent performance in practice. This method was proposed basing on Structural Risk Minimization (SRM) in place of Experiential Risk Minimization (ERM), thus it has good generalization capability. By mapping input data into a high dimensional characteristic space in which an optimal separating hyperplane is built, SVM presents a lot of advantages for resolving the small samples, nonlinear and high dimensional pattern recognition, as well as other machine-learning problems such as function fitting. The simulation results show that the proposed method is effective and feasible.

This paper presents an energy-efficient distributed link scheduling protocol based on vertex coloring method for wireless sensor networks. In the protocol, we perform vertex coloring using maximum degree preferred scheme, and gather network information only by local message exchanges. Thus, our protocol can further reduce the maximum number of timeslots, and avoid long-distance multihop packet forwarding.To improve coloring success rate, two mechanisms are proposed in the protocol. The first is a broadcast guarantee mechanism for increasing broadcast message delivery rate; the second is a conflict processing mechanism for solving coloring conflicts. Simulation results show that the proposed distributed scheduling protocol has better performance than certain centralized scheduling protocols and a classical distributed protocol DRAND in energy efficiency, spatial reuse rate and packet loss rate.

In order to reduce the environmental contact force and make the operation task completed successfully, the robot is frequently required with force perception and active compliance control. Based on the six-axis wrist force sensor measuring, a robot model of surface tracking motion is proposed, and its force control algorithm and experiment are studied. The measurement principle of the six-axis wrist force sensor and the inadequacy of the sensor measuring the six-dimensional force online are introduced firstly. The surface tracking motion model and its coordinate system are established. On this basis, the relationship between the pose adjustment of surface tracking motion and the measuring results of the six-axis wrist force sensor is deduced. At last, the experimental study of the surface tracking robot system that applied the force control algorithm is conducted. The experiment shows that the robot can adjust the current position and orientation in real time according to the six-axis wrist force sensor measuring, which demonstrates the feasibility of the surface tracking motion model and the correctness of the force control algorithm.